BACKGROUND: The effect of antihypertensive medication on cardiovascular diseases is based on results achieved in prospective and controlled clinical studies. Comparable results in clinical practice can be achieved only when the quality of management for subjects with high blood pressure is comparable with the quality achieved in clinical studies. MATERIAL AND METHODS: This study consists of two parts; an attempt to implement an computer-based clinical decision support system for management of hypertension in primary health care in two counties in Norway, and a questionnaire survey among doctors in primary care and specialists in internal medicine in hospitals in other counties. We asked what they expected would be the result of the specific implementation strategy for the computer-based program. The objectives were to evaluate the results of the implementation strategy and to compare these results with expectations expressed in the filled-in questionnaires by doctors without any obligations to the main study. RESULTS: A total of 175 doctors were invited to implement the clinical decision support system. 85% responded; 44% of these, or 37% of the invited doctors, were willing to participate. After 12 months with recurrent visits by one of the authors, only six out of 74 doctors participating in the intervention study still used the program. The questionnaire were completed and returned by 203 doctors, who expected that 55% of the invited doctors would accept the invitation following the implementation strategy used. In general the validity of the information given by the questionnaire was poor and unreliable. INTERPRETATION: We conclude that introduction of a computer-based clinical decision support system is difficult in a busy primary care setting. Availability and simplicity are crucial requirements, and doctors would need financial compensation if they were to use such a system.